Tsumugii/Sunsimiao-1.8B

Partial fine-tune vs Qwen/Qwen1.5-1.8B. Partial change: 83% of the weights changed (attention, lm_head, mlp, norms), in 24 of 24 layers.

Tensors changed: lm_head 1/1, Embeddings 0/1, Norms 4/49, MLP 72/72, Attention 167/168

1 derivatives checked

• mradermacher/Sunsimiao-Qwen1.5-1.8B-full-0.2-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse